Difference between revisions of "Predict"

From CASA Guides
Jump to: navigation, search
(Calculating Visibilities)
Line 2: Line 2:
 
[[Category: Simulations]]
 
[[Category: Simulations]]
  
== Calculating Visibilities ==
+
== Calculating Visibilities ('''obsmode''' = "int") ==
 +
 
 +
Once you have a sky model with appropriate Coordinate System and a list of pointings, you can calculate visibilities.  simobserve can additionally predict from a Component List named '''complist''' (see below).
 +
We only discuss interferometric visibilities here but the process is similar for single dish simulated observations ('''obsmode''' = "sd") -- see [[ACA Guide]]
  
Once you have a sky model with appropriate Coordinate System and a list of pointings, you can calculate visibilities.  simdata can additionally predict from a Component List named '''complist''' (see below).
 
  
 
* '''antennalist''' is a text file containing the antenna locations (see [[Antenna List]] for details)
 
* '''antennalist''' is a text file containing the antenna locations (see [[Antenna List]] for details)
  
* '''refdate''', '''totaltime''' set when you want to observe and for how long.  
+
* '''refdate''', '''hourangle''', and '''totaltime''' set when you want to observe and for how long.  
  
* <font color="green"><tt>simdata</tt> will recenter the track at transit i.e. the maximum altitude</font>.  A long track will see the source rise and set, and placing your model at the appropriate declination can ensure that the source is always low.  More flexibility e.g. for snapshot observations through a range of LST, are currently best achieved by simulating a long track (with image=F if you would like to save time), flagging the timerange near zenith with flagdata, leaving an off-zenith unflagged time range, and then running clean on the ms, either with the clean task, or with simdata (predict=F,noise_thermal="",image=T)
+
* More flexibility of hour angle, e.g. for snapshot observations through a range of LST, are currently best achieved by simulating a long track, flagging the timerange near zenith with <tt>flagdata</tt>, leaving an off-zenith unflagged time range, and then running clean on the ms, either with the clean task, or with simanalyze.
* Mosaic observations will rotate through the mosaic fields during the track defined by '''totaltime''', remaining at each pointing for a duration '''integration'''.  No slew time is inserted.
 
  
* One can interleave a calibrator in direction '''caldirection''' and flux '''calflux''' with one's science pointingsAs of CASA 3.0.2, the pointing file should contain the science mosaic, not the calibratorsimdata will observe all pointings of the pointing file, then the calibrator for the same integration time, then the pointings in the pointing file again, etc, until '''totaltime''' is used up.  Greater flexibility is expected in the future.
+
* Mosaic observations will rotate through the mosaic fields during the track defined by '''totaltime''', remaining at each pointing for a duration '''integration'''.  No slew time is inserted.  '''totaltime''' can be an integer without units, in which case it will cycle through the mosaic that many times.
  
* One can additionally or only create a single dish total power observation.  Set '''sdantlist''' to the file containing the Single Dish antenna location(s)If there is more than one antenna in that file, specify which one to use with '''sdant'''.  
+
* One can interleave a calibrator in direction '''caldirection''' and flux '''calflux''' with one's science pointings.  The pointing file should contain the science mosaic, not the calibratorsimobserve will observe all pointings of the pointing file, then the calibrator for the same integration time, then the pointings in the pointing file again, etc, until '''totaltime''' is used up.
  
* If '''sdantlist''' is set, <tt>simdata</tt> will create $project.sd.ms.  If '''antennalist''' is set, simdata will (also) create $project.ms.
+
* If '''obsmode''' = "sd" is set, <tt>simobserve</tt> will create $project.sd.ms, otherwise it will create $project_$configuration.ms.
  
 
== Component list ==
 
== Component list ==
 
   
 
   
<font color="green">as of May 2010, simdata2 simulations of componentlist ONLY (not modelimage) may not fully work.</font>
 
 
 
One way to get a component list is to deconvolve an image, and another is to create it by hand using the cl tool:
 
One way to get a component list is to deconvolve an image, and another is to create it by hand using the cl tool:
  
Line 34: Line 33:
 
here is a little python program [[File:makecl_py.txt]] that will change this ascii file [[File:my_list.txt]] into a componentlist
 
here is a little python program [[File:makecl_py.txt]] that will change this ascii file [[File:my_list.txt]] into a componentlist
  
<font color="red">warning</font> complist has not been as thoroughly tested as images. email rindebet at nrao.edu with questions or observations.
+
a more complete guide to componentlist simulation is [[Comp list guide]]
  
 
== Graphic Output ==  
 
== Graphic Output ==  

Revision as of 12:22, 4 June 2012

Simulating Observations in CASA

Calculating Visibilities (obsmode = "int")

Once you have a sky model with appropriate Coordinate System and a list of pointings, you can calculate visibilities. simobserve can additionally predict from a Component List named complist (see below). We only discuss interferometric visibilities here but the process is similar for single dish simulated observations (obsmode = "sd") -- see ACA Guide


  • antennalist is a text file containing the antenna locations (see Antenna List for details)
  • refdate, hourangle, and totaltime set when you want to observe and for how long.
  • More flexibility of hour angle, e.g. for snapshot observations through a range of LST, are currently best achieved by simulating a long track, flagging the timerange near zenith with flagdata, leaving an off-zenith unflagged time range, and then running clean on the ms, either with the clean task, or with simanalyze.
  • Mosaic observations will rotate through the mosaic fields during the track defined by totaltime, remaining at each pointing for a duration integration. No slew time is inserted. totaltime can be an integer without units, in which case it will cycle through the mosaic that many times.
  • One can interleave a calibrator in direction caldirection and flux calflux with one's science pointings. The pointing file should contain the science mosaic, not the calibrator. simobserve will observe all pointings of the pointing file, then the calibrator for the same integration time, then the pointings in the pointing file again, etc, until totaltime is used up.
  • If obsmode = "sd" is set, simobserve will create $project.sd.ms, otherwise it will create $project_$configuration.ms.

Component list

One way to get a component list is to deconvolve an image, and another is to create it by hand using the cl tool:

cl.done()
cl.addcomponent(dir="J2000 18h00m00.03s -42d0m0.0s", flux=1.0, freq='672.0GHz')
cl.rename("star672GHz.cl")
complist = "star672GHz.cl"

here is a little python program File:Makecl py.txt that will change this ascii file File:My list.txt into a componentlist

a more complete guide to componentlist simulation is Comp list guide

Graphic Output

If you have graphics turned on, you'll see a diagnostic plot like this one:

N891d.predict.png